sammccall added inline comments.
================
Comment at: clang/lib/Format/TokenAnnotator.cpp:2090
continue;
+ if (Next->isOneOf(tok::star, tok::arrow))
+ continue;
----------------
MyDeveloperDay wrote:
> sammccall wrote:
> > This seems like it's supposed to be handled by the token annotator, and the
> > same cause will result in bugs in other places - why aren't these tokens
> > being marked as TT_OverloadedOperator?
> >
> > I'd guess the deeper fix is in the `kw_operator` case in consumeToken().
> > The way it's written looks suspect, though I'm not an expert on any of this.
> The * from *() is already labelled as a TT_PointerOrRefernce so can't also be
> marked as an Overloaded operator so use that in the loop.
Right, the fact that it's incorrectly labeled as a PointerOrReference is a bug,
is it not possible to fix that instead of working around it here?
In the consumeToken loop, it seems the `*` in `operator *` is labeled as
PointerOrReference by logic that's trying to handle e.g. `StringRef::operator
char*()`. However when the `*` is _immediately_ preceded by `operator`, it's
always an overloaded operator name rather than a pointer, I think?
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D69573/new/
https://reviews.llvm.org/D69573
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
